{"id":"W2950704571","doi":"10.1109/icde.2019.00149","title":"Distributed Discovery of Functional Dependencies","year":2019,"lang":"en","type":"article","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Focus (optics); Distributed computing; Distributed database; Distributed algorithm; Computation; Distributed Computing Environment; Big data; Distributed data store; Data mining; Algorithm","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.001269288,0.00004423152,0.0001163228,0.00006779844,0.00002201259,0.0001102576,0.0003489952,0.00001822118,0.004003143],"category_scores_gemma":[0.0004712596,0.00002775833,0.00005911584,0.0002624188,0.00003845695,0.0006871918,0.0002363315,0.00002871021,0.001543213],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000009510958,"about_ca_system_score_gemma":0.00002282926,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005269159,"about_ca_topic_score_gemma":0.00005524025,"domain_scores_codex":[0.9984438,0.00004882942,0.0002916183,0.0001921987,0.0009379156,0.00008561037],"domain_scores_gemma":[0.9989252,0.0004458526,0.00008790837,0.0004316349,0.00008827663,0.00002112934],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.00006170818,0.0001288262,0.03464805,0.00001489103,0.00003470556,0.000001255593,0.00007266072,0.0005755944,0.001897804,0.7692994,0.1871839,0.006081222],"study_design_scores_gemma":[0.0007405085,0.0001260996,0.5563281,0.00001553502,0.00001554668,0.000002484974,0.007017693,0.001305501,0.004746482,0.1377807,0.2916715,0.000249842],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4661781,0.00002331321,0.4320387,0.001105569,0.0008241624,0.0001734476,0.0003577033,0.00003046954,0.09926857],"genre_scores_gemma":[0.9567003,0.000001775304,0.0001837371,0.000194962,0.00001258353,0.000001363416,0.00005088149,0.000001303539,0.04285312],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6315187,"threshold_uncertainty_score":0.9992342,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.148624552532601,"score_gpt":0.3635099975135841,"score_spread":0.2148854449809832,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}